We report competitive results on object detection and instance segmentation on the COCO dataset using standard models trained from random initialization. The results are no worse than their ImageNet pre-training counterparts even when using the hyper-parameters of the baseline system (Mask R-CNN) that were optimized for fine-tuning pre-trained models, with the sole exception of increasing the number of training iterations so the randomly initialized models may converge. Training from random initialization is surprisingly robust; our results hold even when: (i) using only 10% of the training data, (ii) for deeper and wider models, and (iii) for multiple tasks and metrics. Experiments show that ImageNet pre-training speeds up convergence early in training, but does not necessarily provide regularization or improve final target task accuracy. To push the envelope we demonstrate 50.9 AP on COCO object detection without using any external data---a result on par with the top COCO 2017 competition results that used ImageNet pre-training. These observations challenge the conventional wisdom of ImageNet pre-training for dependent tasks and we expect these discoveries will encourage people to rethink the current de facto paradigm of `pre-training and fine-tuning' in computer vision.

8
下载
关闭预览

相关内容

ImageNet项目是一个用于视觉对象识别软件研究的大型可视化数据库。超过1400万的图像URL被ImageNet手动注释,以指示图片中的对象;在至少一百万个图像中,还提供了边界框。ImageNet包含2万多个类别; [2]一个典型的类别,如“气球”或“草莓”,包含数百个图像。第三方图像URL的注释数据库可以直接从ImageNet免费获得;但是,实际的图像不属于ImageNet。自2010年以来,ImageNet项目每年举办一次软件比赛,即ImageNet大规模视觉识别挑战赛(ILSVRC),软件程序竞相正确分类检测物体和场景。 ImageNet挑战使用了一个“修剪”的1000个非重叠类的列表。2012年在解决ImageNet挑战方面取得了巨大的突破,被广泛认为是2010年的深度学习革命的开始。

Since hardware resources are limited, the objective of training deep learning models is typically to maximize accuracy subject to the time and memory constraints of training and inference. We study the impact of model size in this setting, focusing on Transformer models for NLP tasks that are limited by compute: self-supervised pretraining and high-resource machine translation. We first show that even though smaller Transformer models execute faster per iteration, wider and deeper models converge in significantly fewer steps. Moreover, this acceleration in convergence typically outpaces the additional computational overhead of using larger models. Therefore, the most compute-efficient training strategy is to counterintuitively train extremely large models but stop after a small number of iterations. This leads to an apparent trade-off between the training efficiency of large Transformer models and the inference efficiency of small Transformer models. However, we show that large models are more robust to compression techniques such as quantization and pruning than small models. Consequently, one can get the best of both worlds: heavily compressed, large models achieve higher accuracy than lightly compressed, small models.

0
10
下载
预览

Although pretrained Transformers such as BERT achieve high accuracy on in-distribution examples, do they generalize to new distributions? We systematically measure out-of-distribution (OOD) generalization for various NLP tasks by constructing a new robustness benchmark with realistic distribution shifts. We measure the generalization of previous models including bag-of-words models, ConvNets, and LSTMs, and we show that pretrained Transformers' performance declines are substantially smaller. Pretrained transformers are also more effective at detecting anomalous or OOD examples, while many previous models are frequently worse than chance. We examine which factors affect robustness, finding that larger models are not necessarily more robust, distillation can be harmful, and more diverse pretraining data can enhance robustness. Finally, we show where future work can improve OOD robustness.

0
5
下载
预览

Recent advances have spurred incredible progress in self-supervised pretraining for vision. We investigate what factors may play a role in the utility of these pretraining methods for practitioners. To do this, we evaluate various self-supervised algorithms across a comprehensive array of synthetic datasets and downstream tasks. We prepare a suite of synthetic data that enables an endless supply of annotated images as well as full control over dataset difficulty. Our experiments offer insights into how the utility of self-supervision changes as the number of available labels grows as well as how the utility changes as a function of the downstream task and the properties of the training data. We also find that linear evaluation does not correlate with finetuning performance. Code and data is available at \href{https://www.github.com/princeton-vl/selfstudy}{github.com/princeton-vl/selfstudy}.

0
8
下载
预览

Auto-regressive sequence-to-sequence models with attention mechanism have achieved state-of-the-art performance in many tasks such as machine translation and speech synthesis. These models can be difficult to train. The standard approach, teacher forcing, guides a model with reference output history during training. The problem is that the model is unlikely to recover from its mistakes during inference, where the reference output is replaced by generated output. Several approaches deal with this problem, largely by guiding the model with generated output history. To make training stable, these approaches often require a heuristic schedule or an auxiliary classifier. This paper introduces attention forcing, which guides the model with generated output history and reference attention. This approach can train the model to recover from its mistakes, in a stable fashion, without the need for a schedule or a classifier. In addition, it allows the model to generate output sequences aligned with the references, which can be important for cascaded systems like many speech synthesis systems. Experiments on speech synthesis show that attention forcing yields significant performance gain. Experiments on machine translation show that for tasks where various re-orderings of the output are valid, guiding the model with generated output history is challenging, while guiding the model with reference attention is beneficial.

0
6
下载
预览

Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. Based on this observation, we propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient. We demonstrate the effectiveness of this method on scaling up MobileNets and ResNet. To go even further, we use neural architecture search to design a new baseline network and scale it up to obtain a family of models, called EfficientNets, which achieve much better accuracy and efficiency than previous ConvNets. In particular, our EfficientNet-B7 achieves state-of-the-art 84.4% top-1 / 97.1% top-5 accuracy on ImageNet, while being 8.4x smaller and 6.1x faster on inference than the best existing ConvNet. Our EfficientNets also transfer well and achieve state-of-the-art accuracy on CIFAR-100 (91.7%), Flowers (98.8%), and 3 other transfer learning datasets, with an order of magnitude fewer parameters. Source code is at https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet.

0
3
下载
预览

The current strive towards end-to-end trainable computer vision systems imposes major challenges for the task of visual tracking. In contrast to most other vision problems, tracking requires the learning of a robust target-specific appearance model online, during the inference stage. To be end-to-end trainable, the online learning of the target model thus needs to be embedded in the tracking architecture itself. Due to these difficulties, the popular Siamese paradigm simply predicts a target feature template. However, such a model possesses limited discriminative power due to its inability of integrating background information. We develop an end-to-end tracking architecture, capable of fully exploiting both target and background appearance information for target model prediction. Our architecture is derived from a discriminative learning loss by designing a dedicated optimization process that is capable of predicting a powerful model in only a few iterations. Furthermore, our approach is able to learn key aspects of the discriminative loss itself. The proposed tracker sets a new state-of-the-art on 6 tracking benchmarks, achieving an EAO score of 0.440 on VOT2018, while running at over 40 FPS.

0
6
下载
预览

We present a new approach for pretraining a bi-directional transformer model that provides significant performance gains across a variety of language understanding problems. Our model solves a cloze-style word reconstruction task, where each word is ablated and must be predicted given the rest of the text. Experiments demonstrate large performance gains on GLUE and new state of the art results on NER as well as constituency parsing benchmarks, consistent with the concurrently introduced BERT model. We also present a detailed analysis of a number of factors that contribute to effective pretraining, including data domain and size, model capacity, and variations on the cloze objective.

0
6
下载
预览

The potential of graph convolutional neural networks for the task of zero-shot learning has been demonstrated recently. These models are highly sample efficient as related concepts in the graph structure share statistical strength allowing generalization to new classes when faced with a lack of data. However, knowledge from distant nodes can get diluted when propagating through intermediate nodes, because current approaches to zero-shot learning use graph propagation schemes that perform Laplacian smoothing at each layer. We show that extensive smoothing does not help the task of regressing classifier weights in zero-shot learning. In order to still incorporate information from distant nodes and utilize the graph structure, we propose an Attentive Dense Graph Propagation Module (ADGPM). ADGPM allows us to exploit the hierarchical graph structure of the knowledge graph through additional connections. These connections are added based on a node's relationship to its ancestors and descendants and an attention scheme is further used to weigh their contribution depending on the distance to the node. Finally, we illustrate that finetuning of the feature representation after training the ADGPM leads to considerable improvements. Our method achieves competitive results, outperforming previous zero-shot learning approaches.

0
16
下载
预览

Learning from a few examples remains a key challenge in machine learning. Despite recent advances in important domains such as vision and language, the standard supervised deep learning paradigm does not offer a satisfactory solution for learning new concepts rapidly from little data. In this work, we employ ideas from metric learning based on deep neural features and from recent advances that augment neural networks with external memories. Our framework learns a network that maps a small labelled support set and an unlabelled example to its label, obviating the need for fine-tuning to adapt to new class types. We then define one-shot learning problems on vision (using Omniglot, ImageNet) and language tasks. Our algorithm improves one-shot accuracy on ImageNet from 87.6% to 93.2% and from 88.0% to 93.8% on Omniglot compared to competing approaches. We also demonstrate the usefulness of the same model on language modeling by introducing a one-shot task on the Penn Treebank.

0
9
下载
预览

Zero-shot Learners are models capable of predicting unseen classes. In this work, we propose a Zero-shot Learning approach for text categorization. Our method involves training model on a large corpus of sentences to learn the relationship between a sentence and embedding of sentence's tags. Learning such relationship makes the model generalize to unseen sentences, tags, and even new datasets provided they can be put into same embedding space. The model learns to predict whether a given sentence is related to a tag or not; unlike other classifiers that learn to classify the sentence as one of the possible classes. We propose three different neural networks for the task and report their accuracy on the test set of the dataset used for training them as well as two other standard datasets for which no retraining was done. We show that our models generalize well across new unseen classes in both cases. Although the models do not achieve the accuracy level of the state of the art supervised models, yet it evidently is a step forward towards general intelligence in natural language processing.

0
4
下载
预览
小贴士
相关论文
Zhuohan Li,Eric Wallace,Sheng Shen,Kevin Lin,Kurt Keutzer,Dan Klein,Joseph E. Gonzalez
10+阅读 · 2020年6月23日
Dan Hendrycks,Xiaoyuan Liu,Eric Wallace,Adam Dziedzic,Rishabh Krishnan,Dawn Song
5+阅读 · 2020年4月13日
How Useful is Self-Supervised Pretraining for Visual Tasks?
Alejandro Newell,Jia Deng
8+阅读 · 2020年3月31日
Attention Forcing for Sequence-to-sequence Model Training
Qingyun Dou,Yiting Lu,Joshua Efiong,Mark J. F. Gales
6+阅读 · 2019年9月26日
Mingxing Tan,Quoc V. Le
3+阅读 · 2019年5月28日
Learning Discriminative Model Prediction for Tracking
Goutam Bhat,Martin Danelljan,Luc Van Gool,Radu Timofte
6+阅读 · 2019年4月15日
Alexei Baevski,Sergey Edunov,Yinhan Liu,Luke Zettlemoyer,Michael Auli
6+阅读 · 2019年3月19日
Michael Kampffmeyer,Yinbo Chen,Xiaodan Liang,Hao Wang,Yujia Zhang,Eric P. Xing
16+阅读 · 2018年5月31日
Oriol Vinyals,Charles Blundell,Timothy Lillicrap,Koray Kavukcuoglu,Daan Wierstra
9+阅读 · 2017年12月29日
Pushpankar Kumar Pushp,Muktabh Mayank Srivastava
4+阅读 · 2017年12月23日
相关资讯
Unsupervised Learning via Meta-Learning
CreateAMind
32+阅读 · 2019年1月3日
meta learning 17年:MAML SNAIL
CreateAMind
9+阅读 · 2019年1月2日
A Technical Overview of AI & ML in 2018 & Trends for 2019
待字闺中
10+阅读 · 2018年12月24日
计算机视觉的不同任务
专知
4+阅读 · 2018年8月27日
【SIGIR2018】五篇对抗训练文章
专知
11+阅读 · 2018年7月9日
条件GAN重大改进!cGANs with Projection Discriminator
CreateAMind
6+阅读 · 2018年2月7日
【推荐】YOLO实时目标检测(6fps)
机器学习研究会
17+阅读 · 2017年11月5日
【推荐】深度学习目标检测概览
机器学习研究会
9+阅读 · 2017年9月1日
Auto-Encoding GAN
CreateAMind
5+阅读 · 2017年8月4日
Top